Understanding how the sources of surface water change along river networks is an important challenge, with implications for soil-stream interactions, and our ability to predict hydrological and biogeochemical responses to environmental change. Network-scale patterns of stream water reflect distinct hydrological processes among headwater units, as well as variable contributions from deeper groundwater stores, which may vary nonlinearly with drainage basin size. Here we explore the spatial variability of groundwater inputs to streams, and the corresponding implications for surface water chemistry, during winter baseflow in a boreal river network. The relative contribution of recent and older groundwater was determined using stable isotopes of water (δ<sup>18</sup>O) at 78 locations ranging from small headwaters (0.12 km<sup>2</sup>) to fourth-order streams (68 km<sup>2</sup>) in combination with 79 precipitation and 10 deep groundwater samples. Results from a two end-member mixing model indicate that deeper groundwater inputs increased nonlinearly with drainage area, ranging from ∼20% in smaller headwater subcatchments to 70-80% for catchments with a 10.6 km<sup>2</sup> area or larger. Increases in the groundwater contribution were positively correlated to network-scale patterns in surface stream pH and base cation concentrations and negatively correlated to dissolved organic carbon. These trends in chemical variables are consistent with the production of weathering products and the mineralization of organic matter along groundwater flow paths. Together, the use of stable isotopes and biogeochemical markers illustrate how variation in hydrologic routing and groundwater contributions shape network-scale patterns in stream chemistry as well as patchiness in the relative sensitivity of streams to environmental change and perturbation.
- Mixing model
- Stable isotopes